Princeton University

School of Engineering & Applied Science

Dealing with Variability and Uncertainty on the Road to 80 percent Renewables

Professor William B. Powell, Professor of Operations Research and Financial Engineering, Director, Program in Engineering and Management Systems
Friend Center, 109
Monday, April 14, 2014 - 7:30pm

The Obama administration has set a goal of generating 80 percent of our electricity from renewables, and providing the infrastructure to support one million electric vehicles. This is a daunting challenge for areas such as the mid-Atlantic region of the United States, which has limited access to hydroelectric power and biomass, leaving wind and solar as the primary sources of renewables. In this talk, I will review the issues of variability and uncertainty that arise with wind and solar, highlighting the hopes and realities of using wind and solar to power our grid. I will then describe SMART-ISO, a stochastic, multiscale model of the PJM energy markets and power grid, which has been designed to carefully model the sequencing of decisions and information that arises in energy markets. SMART-ISO has been used to calibrate a robust policy based on a hybrid lookahead model with cost function approximation to handle the variability and uncertainty of energy from wind and solar. We demonstrate the model in a study of off-shore wind.
An issue that comes up in the study of renewables is how to properly analyze them. I will give a light introduction to the proper approach to modeling a stochastic, dynamic system, using as a motivating example the problem of studying the grid with high penetrations of wind and solar.